‘Employment services’ is the term used to refer to a variety of publicly funded, privately delivered operations that are meant to help unemployed people find jobs. It sounds simple, but Australia’s current policy settings in this area are not. And there is good reason to suspect that neither the unemployed nor the broader Australian community are well served by the federal government’s annual $1.8 billion investment in Jobactive.1
Prior to the COVID-19 pandemic, the outcomes achieved by the Jobactive system were mediocre at best: two-thirds of registered clients remain unemployed for at least a year, and approximately 20 per cent of clients have been registered for assistance for at least five years.2 The contractual framework for service providers includes over 3000 pages of regulations. Providers take full advantage of outcomes payments. This involves practices known as ‘creaming’ and ‘parking’: prioritising those who are likely to obtain a new job with little assistance, and underservicing those who are less competitive in the labour market.3
As the longer-term effects of the pandemic begin to become clear, the rate of unemployment in Australia is climbing fast. Given the performance of the employment services sector, redesigning the system will be crucial to our economic recovery and reconstruction. We must challenge the logic of the current system and design a new approach based on evidence-based assumptions about labour market transitions. An enabling employment services system that builds the employability of its clients will also enhance their wellbeing and protect their dignity.
The assumptions of activation
Australia’s employment services have undergone significant transformation since the mid-1970s. The key theoretical principle that has driven the reform agenda and that holds the disparate parts of the employment services sector—including pre-employment programs and specialised support—in some kind of normative order is known as activation.4 The stagflation experienced after the oil shock of the mid-1970s, when both unemployment and inflation increased simultaneously—to the surprise of economists who had not considered this particular coincidence of trends possible—caused a recession that was only resolved by the sweeping economic reforms of the 1980s. During the following decade of strong economic growth, it was increasingly assumed that unemployment was voluntary.5 Recurrent moral panics about ‘dole bludgers’, inflamed by tabloid journalism, entrenched an approach to employment services that by the mid-1990s coupled increasing jobseeking activity with greater surveillance and the threat of financial sanctions, to induce what were perceived as required behavioural changes on the part of unemployed people.
At the simplest level, the activation model holds that behavioural change by individuals will work to reduce unemployment, even in the absence of sufficient numbers of available jobs. While some amount of vocational advice, training and individual case management has been incorporated into the various iterations of employment services implemented over the past few decades, the trinity of activity, surveillance and sanctions is at the core of service design.
While the notion that if people are pushed hard enough they will find work may be intuitively appealing, the activation model is held together by three highly contestable assumptions that actually impede the delivery of effective support.
Activation Assumption 1: The threat of financial penalties will motivate jobseeking behaviours
If individuals are properly motivated they can find work, and the best way to motivate the unemployed is to impose financial sanctions if they are not trying hard enough. Economists call this reducing the disutility of work.6 In an even more Orwellian fashion, policy wonks refer to incentivisation. Politicians have, for more than a quarter of a century, preferred the terms either ‘reciprocal obligation’ or ‘mutual obligation’ to describe the social compact entered into with recipients of unemployment benefits.7
In practice, this means that those on income support, already subsisting on below-poverty-line payments, are subject to financial penalties if they have not undertaken prescribed jobseeking activities in a timely manner. But the aforementioned statistics suggest that this kind of activation has not produced results for a sizeable portion of those seeking work. Further intensification is not, on available evidence, the answer; yet it remains the preferred strategy.
Activation Assumption 2: Constructing a market in employment services will increase efficiency
According to the theory, outsourcing service delivery to a competitive market will promote service innovation and price efficiency. Yet the privatised employment services system is highly regulated and provides little scope to experiment with novel interventions. This is as it must be, if providers are to effectively administer social security legislation through the implantation of activation policies, which essentially involve pressuring unemployed people into submitting endless job applications and undertaking other obligatory tasks that are often tenuously related to getting a job.
Perhaps unsurprisingly, research with those engaged in the system consistently finds them to be bewildered by bureaucracy, chafing under surveillance, and reluctantly undertaking activities that seem irrelevant to their aspirations.8 Research with service provider staff finds that they have little opportunity to exercise professional judgement, and spend increasing amounts of time surveilling and reporting on clients whom they would prefer to help.9
Activation Assumption 3: The economy generates sufficient employment opportunities, efficiently allocating human capital
This assumption is central to the overall logic of the system. There is some evidence that the Australian labour market is reasonably efficient. The OECD calculated that in Australia between 2002 and 2014, about 70 per cent of those retrenched in any given year were able to find employment in the twelve months following.10 However, those who have been retrenched represent only a slice of people looking for work. Indeed, even a conservative estimate of the number of people looking for work at any time over the past decade suggests that there are between three and five unemployed people for each advertised vacancy.11 This ratio increases substantially when the underemployed, and those who have withdrawn from the labour market but would accept work if it were available, are taken into consideration. Yet official unemployment statistics do not include these two categories of people. This means that even when the economy is reasonably robust, it does not generate sufficient employment opportunities for those who want to work. Particular cohorts, including both young and mature-age workers with low levels of education, and people with a disability, are at a particular disadvantage.
The combined outcome of these assumptions is an employment services system that dehumanises its clients. The evidence suggests that the activation framework is more effective in shifting people off welfare by pushing them to detach from the labour market rather than by supporting them into employment.12 Moreover, it does harm, deepening social exclusion rather than enhancing the employability of those facing the greatest disadvantage.13
The assumptions that underpin the current organisation of the employment services sector are shaky at the best of times, and we are not in the best of times.
Cue cataclysms and catalysts for further change
I started writing this chapter just as the April 2020 Labour Force Survey statistics were released, encompassing the month after social distancing restrictions were implemented and the month before JobKeeper payments were to commence.
The accompanying media coverage provided a salutary lesson in the muddled metrics used to measure levels of gainful—that is, directly remunerated—employment in Australia. Although the full impact of the public health measures was yet to be revealed, the fall in the participation rate—the greatest reduction in labour market participation in sixteen years—masked the level of unemployment. As usual, there was little reporting of the rate of underemployment, which rose from 8.8 per cent in March to 13.7 per cent in the month of April.14 However, some media did note that the number of hours worked fell substantially: despite the headline figure of 6.2 per cent unemployment, 2.7 million Australians who could be working were either working less or not at all.15 As I write, the expectation is that even as restrictions start to lift, this figure will continue to rise.
The unemployment crisis precipitated by the pandemic has revealed the flaws in the assumptions that underpin activation. When a contraction in economic activity means that there are fewer available jobs, or when unemployment is structural, which means there is a mismatch between the skills of people in need of work and the occupational requirements of available jobs, incessant job search will not yield outcomes.
COVID-19 is not the only factor changing the world of work. In the face of the greatest spike in unemployment in at least thirty years, other trends affecting the structure and availability of work are likely to continue apace: namely, accelerating technological developments in robotics, artificial intelligence, and cloud and quantum computing; an ageing population; and climate change.16
For some, these trends portend a dystopic future of increasing unemployment and wage stagnation as jobs become increasingly insecure, workers obsolete, and jobless growth the norm. The OECD argues that the same trends also contain the potential to automate unsafe and unsatisfying employment while unleashing creativity and reorganising the nature of paid employment in ways that could reduce barriers to entry for some groups of workers, such as those with caring responsibilities or who might otherwise need to work from home.17 Developments in Australia suggest there is a kernel of truth in both of those predictions, although the effects are likely to be differentiated by occupation and industry, and the risks are more specific to already-disadvantaged jobseekers, particularly those with low levels of education.18
Employment services that enable
The argument for significant reform of employment services is strong, and we must start with a new set of assumptions about what will work to help people find jobs, rather than to reduce government spending on income support. As outlined in the earlier chapter by John Falzon, broader changes to our social security system are needed to address poverty among people who aren’t engaged in paid work, for any reason. There are intrinsic reasons to prefer a system of income support that prevents poverty, but from the perspective of employment services design, it would liberate providers from administering activation. This would open avenues for innovation, and forestall the psychological harm done by the persistent threat of economic insecurity to recipients.
The resulting reform of employment services will require a new set of founding assumptions.
Enabling Assumption 1: Normalising transitions and managing risks will lead to personalised employment services that enhance employability
Over a lifetime, people will transition in and out of paid work at varying levels of intensity as studying commitments and caring responsibilities—even the odd sabbatical—take precedence, or capacity to work is reduced through illness or disability. The reasons why individuals may find it challenging to negotiate these transitions are manifold: economic contraction, industry restructures and regional decline can play a role on the demand side; levels of human capital and social networks have an impact on the supply side. Normalising transitions does not mean that employment services would be inattentive to the risk of social exclusion entailed at points of entry, change and exit, but it would reduce the propensity to punish the unemployed. Instead, the system would work to manage the risks, both structural and individualised, that attend transitions.
Enabling Assumption 2: A redefined public-private partnership in employment services could foster innovation and facilitate transitions
Innovation in complex systems cannot occur in the absence of flexible design principles that allow for localised responsiveness and experimentation. Regulation should function as an enabling constraint. If released from the onerous administration of social security legislation, the public sector could steward the emergence of an employment services system that fosters innovation and rewards services that build people’s employability and prioritise their wellbeing. Key design principles would include responsiveness to the demand-side and supply-side factors that affect levels of unemployment. The aim should be to maximise access to job opportunities through sophisticated job-matching techniques, which for those at a disadvantage in the labour market can improve their chances of securing appropriate, long-term employment.
Enabling Assumption 3: Active and early intervention can mitigate the risk of increasing unemployment
The current JobKeeper scheme, introduced to mitigate the rise in unemployment that accompanied the public health response to COVID-19, is a somewhat blunt example of active and early intervention. However, more direct engagement with industries undergoing restructure, redundancy or reconfiguration, particularly in light of the aforementioned contemporary trends, would be new in the Australian context. Neither human resources departments nor unions are fully engaged in such processes, which means that the most vulnerable workers can fall through the gaps into structural unemployment.
Working in tandem with an improved vocational education system, employment services that specialise in facilitating transitions could fill these gaps by creating transitional labour markets and social enterprises that offer directly remunerated employment, coupled with training, to facilitate the shift between occupations and industries.
Employment services that work
The solution is simple, but not short of revolutionary: we must design a system that builds the capability of unemployed people so that all Australians are supported to navigate the increasingly tumultuous labour market in ways that maximise human dignity and wellbeing.
This will involve reversing the trend to increasing digital surveillance, and jettisoning the logic of activation. Understanding instead that human development is reliant on more than financial incentives, building capability is also important. The market alone will not produce enough jobs at the right level in the right places to minimise unemployment. For these reasons, employment services must:
• personalise support to reduce individual barriers to employability, by providing skills training while building self-esteem and aspiration;
• actively facilitate transitions between occupations in industries under pressure from contemporary trends such as automation and decarbonisation; and
• forge transitional labour markets that can create pathways into secure, mainstream employment.
Australia already has a mixed economy reliant on significant public expenditure and activity, on which we will continue to rely to transition the economy to a sustainable, inclusive yet productive path. In the face of the greatest labour market shock in a century, government can no longer outsource its responsibility to help Australians find good, secure jobs. Now is the time to reinvest the $1.8 billion in employment services that work.